Filter LinkedIn Jobs by Skill and Location (2026 Guide)

Thirdwatch's LinkedIn Jobs Scraper lets recruiter teams and aggregator builders run skill-and-location-targeted queries on a pay-per-job basis — pure HTTP, no login, no browser, full structured output. Built for tech recruiters scoping multi-region searches, sourcing-tool builders surfacing candidate-fit listings, and aggregator pipelines that fork by skill cluster and metro.

Why filter LinkedIn Jobs by skill and location

Skill-and-location targeting is the canonical recruiter-research workflow. According to LinkedIn's 2024 Workforce Report, the platform indexes 14M+ active job listings globally, and over 60% of recruiter searches combine at least one skill keyword with a city or region. For sourcing tools, ATS-enrichment pipelines, and recruiter dashboards, skill-and-location-targeted feeds are the foundation.

The job-to-be-done is structured. A tech recruiter wants 5 skills (Python, React, Kubernetes, Postgres, AWS) × 6 metros (NY, SF, London, Berlin, Bangalore, Singapore) = 30 watchlist combinations refreshed daily. A sourcing-platform team wants to fork their candidate-fit pipeline by skill cluster (frontend vs backend vs ML). An aggregator wants per-region filtered feeds for their job-board partner integrations. All reduce to a query batch + per-query result cap returning structured rows ready for downstream filtering.

How does this compare to the alternatives?

Three options for skill-and-location-targeted LinkedIn jobs data:

LinkedIn Recruiter is the official path but the per-seat cost makes it impractical for automated aggregator pipelines. The LinkedIn Jobs Scraper actor page gives you the same underlying data for daily-monitoring volumes at a fraction of the cost.

How to filter LinkedIn Jobs in 4 steps

Step 1: How do I authenticate against Apify?

Sign in at apify.com (free tier, no credit card), open Settings → Integrations, and copy your personal API token. Every example below assumes the token is in APIFY_TOKEN:

export APIFY_TOKEN="apify_api_xxxxxxxxxxxxxxxx"

Step 2: How do I compose a skill × location batch?

Pass an array of pre-composed query strings combining skill and location tokens.

import os, requests, pandas as pd
from itertools import product

ACTOR = "thirdwatch~linkedin-jobs-scraper"
TOKEN = os.environ["APIFY_TOKEN"]

SKILLS = ["senior python engineer", "senior react engineer",
          "senior kubernetes engineer", "senior data scientist",
          "senior machine learning engineer"]
METROS = ["New York", "San Francisco", "London", "Berlin",
          "Bangalore", "Singapore", "remote"]

queries = [f"{s} {m}" for s, m in product(SKILLS, METROS)]
print(f"{len(queries)} skill × location queries")

resp = requests.post(
    f"https://api.apify.com/v2/acts/{ACTOR}/run-sync-get-dataset-items",
    params={"token": TOKEN},
    json={"queries": queries, "maxResults": 50},
    timeout=3600,
)
df = pd.DataFrame(resp.json())
print(f"{len(df)} jobs across {df.location.nunique()} locations")

5 skills × 7 metros = 35 queries × 50 = up to 1,750 jobs — well within budget for a daily refresh.

Step 3: How do I post-filter for skill precision?

LinkedIn's relevance ranking is loose — match skill keywords against description for higher precision.

SKILL_KEYWORDS = {
    "python": ["python", "django", "flask", "fastapi"],
    "react": ["react", "next.js", "redux"],
    "kubernetes": ["kubernetes", "k8s", "helm", "kustomize"],
    "ml": ["machine learning", "pytorch", "tensorflow", "sklearn"],
}

def has_skill(description, skill):
    if not isinstance(description, str):
        return False
    return any(kw in description.lower() for kw in SKILL_KEYWORDS[skill])

for skill in SKILL_KEYWORDS:
    df[f"is_{skill}"] = df.description.apply(lambda d: has_skill(d, skill))

precision = df[df.is_python].copy()
print(f"{len(precision)} python-confirmed jobs (vs {(df.title.str.lower().str.contains('python')).sum()} title-matched)")

Description-keyword matching catches 90%+ relevant roles vs ~70% from title-only matching. The 20-point gain compounds at scale.

Step 4: How do I split remote vs onsite?

LinkedIn's location field carries the canonical remote signal — postings tagged remote include "Remote" in the location string.

df["is_remote"] = df.location.str.contains("Remote", case=False, na=False)
df["is_hybrid"] = df.description.str.contains(
    r"hybrid|in[- ]office \d+ days|days? per week", case=False, na=False, regex=True
)
df["is_onsite"] = ~df.is_remote & ~df.is_hybrid

split = df.groupby(["is_remote", "is_hybrid", "is_onsite"]).size()
print(split)

Splitting by work-mode lets recruiters route remote-eligible postings to remote-first sourcing pipelines and onsite postings to local recruiters. Hybrid-mode detection requires description-keyword matching since LinkedIn doesn't expose a structured hybrid flag.

Sample output

A single LinkedIn job record with skill-relevant fields populated looks like this. Five rows weigh ~6 KB.

{
  "title": "Senior Python Engineer",
  "company_name": "Stripe",
  "location": "San Francisco, CA (Remote)",
  "experience_level": "Mid-Senior level",
  "job_type": "Full-time",
  "skills": ["Python", "Django", "PostgreSQL", "Kubernetes"],
  "description": "We are looking for a Senior Python Engineer to join our team...",
  "applicant_count": "200+ applicants",
  "is_easy_apply": true,
  "posted_at": "2026-04-25",
  "apply_url": "https://www.linkedin.com/jobs/view/123456/"
}

skills is the killer field for skill-filter pipelines — it's LinkedIn's parsed skill-tag array, separate from description text. For cross-query dedup, apply_url is canonical. posted_at lets you isolate truly-new postings vs re-listings. is_easy_apply: true indicates LinkedIn-hosted apply flow vs employer ATS redirect.

Common pitfalls

Three things go wrong in skill-and-location filtering pipelines. Skill-tag sparseness — only ~50% of postings populate the skills array; for the rest, fall back to description-keyword matching. Location-string drift — LinkedIn's location formats vary ("San Francisco, CA", "San Francisco Bay Area", "Greater San Francisco"); for clean grouping, normalise via a metro-mapping table rather than exact-string matching. Boolean operator handling — LinkedIn's web search supports basic boolean (AND, OR, quoted phrases) but the actor's keyword passthrough varies; for highest precision, compose multiple separate queries rather than relying on complex Boolean strings.

Thirdwatch's actor uses a lightweight HTTP path so you pay only for the data, not for proxy or compute overhead. Pair LinkedIn Jobs with LinkedIn Profile Scraper to enrich shortlisted candidates with profile data once a target listing is identified. A fourth subtle issue worth flagging: LinkedIn occasionally surfaces postings with stale posted_at values (the field reflects original-posting date, not relisting date) — for recency-sensitive sourcing, supplement posted_at with the scrape timestamp and treat the larger of the two as your "actively-listed since" anchor. A fifth pattern unique to skill-and-location pipelines: certain mega-employers (Google, Amazon, Meta) post the same role across multiple metros simultaneously to scope geographic flexibility, which inflates per-metro role counts; dedupe on (company_name, title, posted_at) before counting per-metro to avoid this distortion. A sixth pitfall: some emerging skill keywords (e.g., very new framework names) return sparse LinkedIn results despite real demand — for cutting-edge skills, supplement LinkedIn searches with GitHub Jobs and Wellfound queries to capture the early-adopter segment that LinkedIn under-indexes. A seventh and final pattern worth noting for international recruiter pipelines: LinkedIn's location matching is materially weaker for non-English city names — queries like senior python engineer München return fewer results than senior python engineer Munich, even though both refer to the same metro. For European, LATAM, and APAC pipelines, always use the canonical English city name in queries and post-process the location field for native-language address strings rather than relying on LinkedIn to match across language variants.

Related use cases

• Scrape LinkedIn Jobs without login
• Build a LinkedIn Jobs aggregator with Apify
• Track LinkedIn hiring velocity by company
• The complete guide to scraping job boards
• All Thirdwatch use-case guides